cpu.rs: Refactor and Fix ARM/Aarch64 CPU features handling. #1100
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
briansmith
force-pushed
the
b/aarch64-static-feature-detection
branch
from
9943ce8 to
66d8c5e
Compare
Presently on aarch64-apple-* `GFp_armcap_P` is always zero. That's wrong; the assembly language code needs it to be set correctly, or else the most optimized code paths (NEON and/or SHA-2 extensions) will never be chosen. Refactor the code so that `GFp_armcap_P` is set correctly, and to make it easier to understand and maintain. This will enable more optimized implementations on aarch64-apple-* targets, whereas before the lowest common denominator implementations were being used for any features that did the feature detection in assembly language code instead of Rust. Move the definition of `GFp_armcap_P` to Rust so wouldn't have to keep C and Rust code for it in sync. Remove the fallback definitions of `GFp_armcap_P` that use the ".comm"; they would always be set to zero if they were ever used, which wouldn't (necessarily) match the static feature set. Removing them makes it clearer that those definitions aren't used.
briansmith
force-pushed
the
b/aarch64-static-feature-detection
branch
from
66d8c5e to
0f7a0c0
Compare
This was referenced Nov 18, 2020
This was referenced Dec 3, 2020
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Presently on aarch64-apple-*
GFp_armcap_Pis always zero. That's wrong; theassembly language code needs it to be set correctly, or else the most optimized
code paths (NEON and/or SHA-2 extensions) will never be chosen.
Refactor the code so that
GFp_armcap_Pis set correctly, and to make it easier tounderstand and maintain.
This will enable more optimized implementations on aarch64-apple-* targets, whereas
before the lowest common denominator implementations were being used for any
features that did the feature detection in assembly language code instead of Rust.
Move the definition of
GFp_armcap_Pto Rust so wouldn't have to keep C and Rustcode for it in sync. Remove the fallback definitions of
GFp_armcap_Pthat use the".comm"; they would always be set to zero if they were ever used, which wouldn't
(necessarily) match the static feature set. Removing them makes it clearer that
those definitions aren't used.